Another bug using RH7.0 version of g++...

Fons Rademakers Fons.Rademakers@cern.ch
Tue Oct 3 05:21:00 GMT 2000


Hhmmm, RH7.0 should never ship with an "experimental" compiler. If after
a stabilzation period the gcc maintainers still don't consider the compiler
stable (even with all stabilizing features added) it should not have been
put on an "official" distribution. The problem is that I as application
provider must now support this "exprimental" compiler (which means introducing
all kind of work arounds in the source and/or an overall lower optimization
level). Or should I tell all my users to who installed RH7.0 to first install
gcc 2.95.2 before compiling my app?

Anyway, as I reported later yesterday on gcc-bugs, the bugs I found in RH 7.0's
version of gcc are still there in the current CVS version so the bug reports
belong on gcc-bugs. I hope somebody gets around fixing them.

Besides the compiler crashes I also noted numerical instabilities when
compiling with -O2 which cause my program to crash. With -O things are fine.


Cheers, Fons.



Alexandre Oliva wrote:
> 
> On Oct  2, 2000, llewelly@edevnull.com wrote:
> 
> > Fons Rademakers <Fons.Rademakers@cern.ch> writes:
> >> Did anybody test this compiler before making it the default of RH7.0?!?
> 
> > Please note that gcc 2.96 is an experimental compiler, *not* a
> >   released compiler.
> 
> Actually, Red Hat Linux 7 shipped with a version of GCC that started
> as a CVS snapshot, but that undergone a long stabilization process,
> with lots of bug-fixes that were contributed back to the official GCC
> repository.  It is definitely not experimental, but it isn't an
> official release of GCC either, in the sense that it was not released
> by someone appointed by the FSF or the GCC Steering Committee.
> 
> For that reason, this is not the best place to report bugs on that
> compiler, unless the bug is known to be present in some official
> release of GCC, or in a recent development snapshot.  It's unfortunate
> that the compiler Red Hat shipped tells users to report bugs here; the
> FSF is not responsible for bugs in Red Hat products.  Instead, it
> would have been more appropriate to point to bugzilla.redhat.com.
> 
> --
> Alexandre Oliva   Enjoy Guarana', see http://www.ic.unicamp.br/~oliva/
> Red Hat GCC Developer                  aoliva@{cygnus.com, redhat.com}
> CS PhD student at IC-Unicamp        oliva@{lsd.ic.unicamp.br, gnu.org}
> Free Software Evangelist    *Please* write to mailing lists, not to me

-- 
Org:    CERN, European Laboratory for Particle Physics.
Mail:   1211 Geneve 23, Switzerland
E-Mail: Fons.Rademakers@cern.ch              Phone: +41 22 7679248
WWW:    http://root.cern.ch/~rdm/            Fax:   +41 22 7677910


More information about the Gcc-bugs mailing list